Xing Haiqun, Hao Zhixin, Zhu Wenjia, Sun Dehui, Ding Jie, Zhang Hui, Liu Yu, Huo Li
Department of Nuclear Medicine, Peking Union Medical College Hospital, Dongcheng District, Chinese Academy of Medical Science, Peking Union Medical College, No.1 Shuaifuyuan, Beijing, 100730, China.
Beijing Key Laboratory of Molecular Targeted Diagnosis and Therapy in Nuclear Medicine, Beijing, 100730, China.
EJNMMI Res. 2021 Feb 25;11(1):19. doi: 10.1186/s13550-021-00760-3.
To develop and validate a machine learning model based on radiomic features derived from F-fluorodeoxyglucose (F-FDG) positron emission tomography/computed tomography (PET/CT) images to preoperatively predict the pathological grade in patients with pancreatic ductal adenocarcinoma (PDAC).
A total of 149 patients (83 men, 66 women, mean age 61 years old) with pathologically proven PDAC and a preoperative F-FDG PET/CT scan between May 2009 and January 2016 were included in this retrospective study. The cohort of patients was divided into two separate groups for the training (99 patients) and validation (50 patients) in chronological order. Radiomics features were extracted from PET/CT images using Pyradiomics implemented in Python, and the XGBoost algorithm was used to build a prediction model. Conventional PET parameters, including standardized uptake value, metabolic tumor volume, and total lesion glycolysis, were also measured. The quality of the proposed model was appraised by means of receiver operating characteristics (ROC) and areas under the ROC curve (AUC).
The prediction model based on a twelve-feature-combined radiomics signature could stratify PDAC patients into grade 1 and grade 2/3 groups with AUC of 0.994 in the training set and 0.921 in the validation set.
The model developed is capable of predicting pathological differentiation grade of PDAC based on preoperative F-FDG PET/CT radiomics features.
开发并验证一种基于从氟脱氧葡萄糖（F-FDG）正电子发射断层扫描/计算机断层扫描（PET/CT）图像中提取的放射组学特征的机器学习模型，以术前预测胰腺导管腺癌（PDAC）患者的病理分级。
本回顾性研究纳入了2009年5月至2016年1月期间149例经病理证实为PDAC且术前行F-FDG PET/CT扫描的患者（83例男性，66例女性，平均年龄61岁）。按时间顺序将患者队列分为两个独立的组，分别用于训练（99例患者）和验证（50例患者）。使用Python中实现的Pyradiomics从PET/CT图像中提取放射组学特征，并使用XGBoost算法构建预测模型。还测量了包括标准化摄取值、代谢肿瘤体积和总病变糖酵解在内的传统PET参数。通过受试者操作特征（ROC）和ROC曲线下面积（AUC）评估所提出模型的质量。
基于十二个特征组合的放射组学特征的预测模型可将PDAC患者分为1级和2/3级组，训练集的AUC为0.994，验证集的AUC为0.921。
所开发的模型能够基于术前F-FDG PET/CT放射组学特征预测PDAC的病理分化程度。
